• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

AMD Ryzen Zen 5 XD releases November 7th

Draugoth

Draugoth

Gold Member
ryzenx3d9000series.png

AMD is officially announcing its first next-generation X3D desktop processors today, based on its latest Zen 5 architecture.

Thanks to other leaks, we’re expecting to hear more about the 9800X3D soon, which will likely be the first next-gen X3D processor to launch on November 7th, with eight cores. Rumors suggest it will include a 4.7GHz base clock, with a boost clock of 5.2GHz or higher.
Click to expand...

Reviews
​

AMD Ryzen 9000 now cheaper​

AMD is also announcing a price cut for the Ryzen 9000X non-3D series. Early adopters may be disappointed to hear that the Zen 5 desktop CPUs are now $30 to $50 cheaper than they were two months ago. The official price cuts are as follows:

AMD Ryzen 9000 Deals (affiliate links)​

 
Last edited:
CJ_75

CJ_75

Member
Will this have any benefit over the Ryzen 9 7900X in a 4090 gaming rig playing in 4K? It wasn’t for the 7800X3D.
 
GHG

GHG

Gold Member
SolidQ said:
Also 9950X + 3090
faa3fb795a987aeca606afced15af4ba.png



14900 + 4090 + 8400С36
edf747911ffb8edb53da591414b0a3e4.png
Click to expand...

Those are run at different resolutions to the 9800x3d benchmark though.

I'm not familiar with this benchmark, does the resolution affect the final number being presented?
 
Celcius

Celcius

°Temp. member
videocardz.com

AMD Ryzen 9000X3D officially arrives November 7th, 9000X series now $30 to $50 cheaper - VideoCardz.com

AMD teases Ryzen 7 9800X3D on November 7th, 9000X gets an official price cut for the Holiday AMD today confirms the release date of the first X3D processor. Just three days before Intel lifts the embargo on Core Ultra 200K reviews and sales, AMD is teasing the successor to its X3D lineup, based...
videocardz.com videocardz.com

They're also dropping the price on zen 5 non-X3D
I'm definitely waiting for reviews on the 9800X3D before deciding how to upgrade this fall...
 
D

dave_d

Member
GHG said:
I'm not familiar with this benchmark, does the resolution affect the final number being presented?
Click to expand...
I think the deal is that you use a lower resolution so the speed of the video card won't affect the result. (Pretty much any cpu specific benchmark I've seen they run at lower resolutions because of that. That is opposed to say full 4k where most of the work is being done by the video card.)
 
GHG

GHG

Gold Member
dave_d said:
I think the deal is that you use a lower resolution so the speed of the video card won't affect the result. (Pretty much any cpu specific benchmark I've seen they run at lower resolutions because of that. That is opposed to say full 4k where most of the work is being done by the video card.)
Click to expand...

I know that, it's the benchmark result number that it's spitting out that I'm questioning.

Some benchmarks posted in this thread are 720p whereas others are 1080p.
 
ap_puff

ap_puff

Member
SolidQ said:
Remastered. REIMAGINED


There is answer in posts
Click to expand...

maybe I'm reading into it too much but does that triple layer profile on the X3D design lend some credibility to the double-stacked 3d vcache theory that High Yield had? Also seeing some leaked prices showing it's gonna be $50 more expensive than the 7800X3D launch. If it's got more expensive packaging that would make sense
 
Last edited:
ap_puff

ap_puff

Member
SolidQ said:
Seems so here is answer also
5ff281905b840e7bcb9fe9a6f1963448.png
Click to expand...
Oh interesting, so instead of doing the cache die on top, they're going to do it on the bottom? I guess if that means you get the compute area closer to the IHS that would help with heat dissipation. But what about voltages? I thought one of the reasons why 3dcache was clocked lower was lower voltage tolerances, but if you're doing the cache on the bottom then wouldn't it be exposed to higher voltage?
 
SolidQ

SolidQ

Gold Member
ap_puff said:
But what about voltages? I thought one of the reasons why 3dcache was clocked lower was lower voltage tolerances, but if you're doing the cache on the bottom then wouldn't it be exposed to higher voltage?
Click to expand...
Need wait to see how it's going. More leaks soon coming or two weeks from release. Also waiting Deep Dive from Chips and Cheese
 
b0uncyfr0

b0uncyfr0

Member
Hmm definitely skipping this one. My 5800x3d will hold out.

But do we have any news on DDR6? Knowing AMD, testing and first adopters will suffer as per usual. Maybe AM6 will have it.
 
S

Schnauzer

Member
Sounds like this will be a 2-3% performance increase. If they had something worth showing, they would have played their cards before the Intel release not afterwards.
 
marquimvfs

marquimvfs

Member
ap_puff said:
Oh interesting, so instead of doing the cache die on top, they're going to do it on the bottom? I guess if that means you get the compute area closer to the IHS that would help with heat dissipation. But what about voltages? I thought one of the reasons why 3dcache was clocked lower was lower voltage tolerances, but if you're doing the cache on the bottom then wouldn't it be exposed to higher voltage?
Click to expand...
That's not how it works, voltage are provided by Motherboard's VRM at a stable level, it won’t get significantly higher or lower if you place things differently. But being under the cores die could force the clocks to be even lower to keep it's temperature in check.
 
ap_puff

ap_puff

Member
marquimvfs said:
That's not how it works, voltage are provided by Motherboard's VRM at a stable level, it won’t get significantly higher or lower if you place things differently. But being under the cores die could force the clocks to be even lower to keep it's temperature in check.
Click to expand...
No, voltage drops as you run current through resistors. Cache being on top of the compute die means that it will be exposed to lower voltage as the current has to go from the VRM->traces->TSVs/compute die->cache. Maybe it's only a few mv but it makes a difference when the 3d cache is voltage sensitive to begin with. Btw the cache is drawing much less power than the CPU cores (hence cooler), and having the cache chiplet underneath actually improves cooling as right now the cache chiplet only covers the L3 of the base chip + a small portion of the cores+L2 cache, with the "empty space" filled with structural silicon. Moving the core chiplet up top would actually give a signifcant improvement in cooling as now it has greater direct surface area contact with the IHS/interposer.
 
marquimvfs

marquimvfs

Member
ap_puff said:
Moving the core chiplet up top would actually give a signifcant improvement in cooling as now it has greater direct surface area contact with the IHS/interposer.
Click to expand...
ap_puff said:
No, voltage drops as you run current through resistors.
Click to expand...
ap_puff said:
Cache being on top of the compute die means that it will be exposed to lower voltage as the current has to go from the VRM->traces->TSVs/compute die->cache.
Click to expand...
ap_puff said:
having the cache chiplet underneath actually improves cooling
Click to expand...
well, maybe I'm in the wrong here, but didn't the interposer feed both the CPU and cache in parallel? The top chiplet still had direct connection with the interposer, no? If I'm right, the conductor distance wasn't long enough to cause any significant resistance, decreasing it's size shouldn't increase the voltage at a significant level.

Also, I get that moving cores die would greatly improve general heat dissipation, but, and again, I might be wrong here, as much as cache got heat from cores above in the old design, being in direct contact with the ihs maked it's dissipation better than if being below. Now, being under cores, as much as there's no heat from below, the cache dissipation is worse than before. What I don't know is that with less heat from below, but also more heat fr above and less dissipation capabilities, if the cache will, in fact, be cooler. If I'm not mistaken, it will be hotter.
 
Last edited:
Celcius

Celcius

°Temp. member
Here's what I get with my 10700k + rtx 3090 at 1080p lowest settings:

EzG2Nqn.jpeg


This cpu upgrade is going to be huge for me. Folks with the same videocard & settings but newer cpus are getting literally double the score so I'm being held back big time.
 
Last edited:
DonkeyPunchJr

DonkeyPunchJr

World’s Biggest Weeb
8% faster at gaming vs 7800X3D. Wonder what resolution that’s at. Seems like a pretty small generational improvement compared to the 5800X3D -> 7800X3D improvement.

www.techspot.com

AMD Ryzen 7 7800X3D vs 5800X3D

If you are building or upgrading a Ryzen gaming PC, should you go with the more affordable 5800X3D, or go all out with the 7800X3D? Today's review...
www.techspot.com www.techspot.com
 
You must log in or register to reply here.

Similar threads

Ryzen 9000 series official prices
Hardware 2
55
3K
Insane Metal replied
  • Locked
Zen 5 Official Pricing Announced
News Hardware
9
911
Insane Metal replied
AMD Ryzen 9000 “Zen 5” Desktop CPU Prices Leaked: 9950X $499, 9900X $399, 9700X $299, 9600X $229
Rumor 2 3 4
178
10K
The Mad Draklor replied
AMD’s Zen 5 processors will arrive on July 31st.
News 2
55
4K
simpatico replied
AMD presents Zen 5
Business 2 3 4 5
212
16K
Celcius replied
Top Bottom